Potatoes and eggplants leaf disease detection using deep learning approach

Abstract

This study explores the application of deep learning techniques for the detection of leaf diseases in potatoes and eggplants, which are significant for moving forward agrarian efficiency and nourishment security. Conventional strategies of malady discovery, which depend on manual assessment, are time-consuming and regularly wrong. In reaction, this inquiry utilizes convolutional neural networks (CNNs), leveraging huge datasets of leaf pictures to prepare models that can consequently identify and classify maladies. The technique incorporates information collection from both field tests and freely accessible datasets, followed by the usage of a few CNN designs like VGG16, VGG19, and MobileNetV2. The models were assessed based on their precision in recognizing different infection sorts, with a specific focus on moving forward early location capabilities. The discoveries illustrate that profound learning models altogether outflank conventional strategies, advertising a quick, dependable, and adaptable arrangement for early malady location in potato and eggplant crops. This progression has the potential to not only upgrade trim administration but also diminish pesticide utilization, hence contributing to economical rural housing. Future work will center on refining these models and sending them to real-world rural settings to approve their commonsense adequacy.